PATIENT REPORTED OUTCOMES

1. ICTR Clinical Registry Workshop, 10 November 2010 PATIENT REPORTED OUTCOMES Albert W. Wu, MD, MPH Joseph Finkelstein, MD, PhD, MA, CCRP

2. Dr. Albert Wu is Professor of Health Policy and Management, with joint appointments in Epidemiology and in International Health in the Bloomberg School of Public Health, and Medicine, and Surgery in the School of Medicine. Course Faculty

3. Dr. Joseph Finkelstein is Associate Professor of Medicine and Director, Chronic Disease Informatics Program in the Johns Hopkins School of Medicine Course Faculty

4. Overview 1.       Importance of patient perspective 2.       Definitions 3.       Key concepts to measure as predictors and outcomes in clinical registries 4.       Measurement of PROs 5.       Class exercise: complete & score the SF-36 6.       Criteria for selecting a tool for a specific registry 7. Form-Builder: a module for PRO data collection

5. Some Questions Cannot Be Answered Without Asking the Patient • The main objective of much of health care is improving how patient feels and functions • Reduction in pain (hip replacement) • Improved functioning (cataract extraction) • Patient is best judge • Patient best observer of some events and health outcomes (complications)

6. Types of Information Captured in Data Used for CER

7. What is a PRO? • “Patient-reported outcomes represent the patient’s report of a health condition and its treatment” (Acquadro et al. Value in Health 2003;5:522-531) • “Any report coming directly from patients (i.e., study subjects) about a health condition and its treatment” (FDA Draft PRO Guidance)

8. Categories of Patient Outcomes Patient Outcomes Assessment Sources and Examples Clinician-Reported Patient-Reported Physiological Caregiver-Reported For example, Global impressions Observations & tests of function For example, FEV1 HbA1c Tumor size For example, Dependency Functional status Global Impression Functional status Well-being Symptoms HRQL Satisfaction with TX Treatment adherence Utility/preference-based measures Source: Acquadro et al. Value in Health 2003;5:522-531

9. Conventional Clinical Measures • 1. Mortality • 2. Disease or treatment complications • 3. Pathology • 4. Physiologic or lab abnormalities • 5. Deformity • 6. Signs and symptoms

10. Outcomes from Different Perspectives • Clinical Perspective • Patient Perspective • Subjective health status • Quality of life • Satisfaction • Societal Perspective • Utilization • Cost

11. Definitions (Apologies…) Imprecise • Many terms used interchangeably • Health • (Subjective) health status • Functional status • Quality of life • Health-related quality of life

12. PROs HRQoL Satisfaction Utility Behavior Symptoms Utilization

13. The Six “D’s” of Outcomes Research • Death • Disease • Disability • Discomfort • Dissatisfaction • Dollars

14. World Health Organization definition of Health (1948) • "a state of complete physical, mental, and social well-being, and not merely the absence of disease and infirmity".

15. Health-related Quality of Life • ….encompasses several aspects of health that are directly experienced by the person including physical functioning, social and role functioning, mental health, general health perceptions.

16. Relationship of Pathophysiology to Subjective Health and QOL Patho- Symptoms Physical/Mental QOL physiology Health Wilson & Cleary, JAMA

17. How do you Measure HRQOL?

19. Do you feel sad?

20. Do you wake up in the middle of the night? Do you feel sad?

21. Do you feel worthless? Do you wake up in the middle of the night? Do you feel sad?

22. Measuring HRQOL • No standard scale, need to specify what we want to measure • Assemble several indicators which approximate the concept • Create scale scores by combining responses to questions

23. How Do You Measure HRQOL? • An infinite number of indicators would fully represent the concept • To be practical, assemble several indicators which approximate the concept • Create scale scores by combining responses to questions

25. Medical Outcomes Study Conceptual Framework Process of Care Outcome of Care System Characteristics Clinical End Points Provider Technical Style Provider Characteristics Functional Status Patient Characteristics Providers’ Interactional Style General Well-Being Satisfaction with Care

27. Language • Plain language • 8th grade reading level • Short questions • English

28. Questions • Reports and ratings • Single barrelled (no “and”s) • Positive and negatively worded • Redundant • Range of ability

29. Response Categories • Adjective rating response scale • Ordered responses • Assigned numerical values

30. Content and administration • Self-administered, also phone or interview • Standardized (different) 3-6 point response scales • Time frame last 4 weeks

31. Scoring • Response are assigned numerical values • Scores summed for each dimension • 8 Dimensions scored on 0-100 scale • 2 Summary Score • Physical Component Score • Mental Component Score

32. Scoring Example: MHI-5 • 9b. Have you been a very nervous person? • 9c. Have you felt so down in the dumps that nothing could cheer you up? • 9d. Have you felt calm and peaceful? • 9f. Have you felt downhearted and blue? • 9h. Have you been a happy person?

33. Checkone answer All of the Time Most of the Time A Good Bit of the Time Some of the Time A Little of the Time None of the Time

34. Scaling and scoring - • Sum of item scores • Recode and reverse • Linear conversion to 0-100 scale

35. Item Scoring • Items 9b, 9f, 9c - use precoded values • Items 9d and 9h - require recoding prior to computation of the scale score • All of the Time: 1 = 6 • Most of the Time 2 = 5 • A Good Bit of the Time 3 = 4 • Some of the Time 4 = 3 • A Little of the Time 5 = 2 • None of the Time 6 = 1

36. Scale Scoring • Recode and reverse items • Compute sum of recoded items scores • 5 items, 6 response categories: Lowest possible score 5, highest 30 (range 25) • Transformation of raw summated score: • [(Raw scale score - Lowest possible score) / possible score range] x 100 • e.g. Mental health score of 21 • [(21 - 5)/25] x 100 = 64

37. Mental Health Scores in General US Population (n=2459)

39. T-Scores for the SF-36 • Mean = 50 • SD = 10

40. Generic vs Disease-Specific Generic Measures Can be used across populations Generally better-tested Disease (Treatment, Population, Study) Specific Measures At least theoretically more sensitive to difference or changes

41. Q: “Should I use the SF-36 or the SIP in my study of liver transplant patients?”

42. A: • What is your research question? • Who are the patients you are studying? • What do you anticipate will happen?

43. Selection of a Health Status Measure • Appropriateness: of the measure to the question or issue of concern • Evidence in relevant populations of: • Reliability, Validity, Responsiveness • Practical considerations

44. Appropriateness: • Of the measure to the question or issue of concern • Correspondence between the content of the measure and goals of the study • Always examine the questionnaire itself • Do scales go into sufficient depth? • Range in study sample vs the instrument • Level of aggregation of scores

45. Evidence in relevant populations • Reliability • Validity • Face validity • Content • Construct • Responsiveness • Pilot test?

46. Practical considerations: • Mode of administration • Time to administer • Language • Respondent burden • Availability of supporting materials

47. Example

48. Conclusion • PRO important predictors and outcomes in clinical research studies • Key concepts to measure include health related quality of life, symptoms, satisfaction, adherence • Measurement relies on questions and scales as indicators of latent constructs of interest • SF-36 and Form-builder as examples • Selection based on research questions, evidence of usefulness in your population, practical considerations • Criteria for selecting a tool for a specific registry

49. “Listen to the patient: He is telling you the diagnosis” - William Osler